Abstract
BACKGROUND: Surgical stress has been shown to result in immune disturbance. Neuraxial anesthesia (NA) has long been hypothesized to blunt undesired surgical insults and thus limit immune compromise and improve surgical outcomes. We hypothesized that NA would decrease postoperative infectious complications compared with general anesthesia (GA) among knee arthroplasty patients. METHODS: We studied the American College of Surgeons National Surgical Quality Improvement Program database from 2005 to 2010. There were 16,555 patients included in our final cohort, with 9167 patients receiving GA and 7388 patients receiving spinal or epidural anesthesia.. Outcomes of interest included infection-related 30-day postoperative complications, including surgical site-related infections, pneumonia, urinary tract infection, sepsis, septic shock, and a composite end point of any systemic infection. Multivariable logistic regression was performed to test for effect of anesthesia type while adjusting for the influence of preexisting comorbidities. RESULTS: The overall mortality was 0.24% and 0.15% among NA and GA subjects, respectively (P = 0.214). NA subjects had fewer unadjusted incidences of pneumonia (P = 0.035) and composite systemic infection (P = 0.006). After risk adjustment for preexisting comorbidities, NA was associated with lower odds of pneumonia (odds ratio = 0.51 [95% confidence interval, 0.29-0.90]) and lower odds of composite systemic infection (odds ratio = 0.77 [95% confidence interval, 0.64-0.92]). CONCLUSIONS: Our study suggested that NA was associated with lower adjusted odds of both pneumonia and a composite outcome of any systemic infectious complication within 30 days of surgery compared with GA.